Configuring endpoint evaluation
To properly log scoring requests for quality, model evaluation uses endpoints. Fairness and drift evaluation use a payload logging endpoint. Quality evaluation uses the feedback endpoint. You can generate a code snippet for the payload and feedback endpoints and for debiased transactions so that you can integrate them into your application.
Steps
- On the Evaluations window, click Configure monitors.
- In the navigation panel, click Endpoints.
- In the Information panel, click Endpoints.
- From the Endpoint list, choose the type of endpoint: Payload logging, Feedback logging, or Debiased transactions.
- From the Code language list, choose the type of code: cURL, Java, or Python.
- To copy the code snippet, click the Copy to clipboard icon.
You can also click Upload feedback data to upload feedback data with a CSV file. For production models, you can click Upload payload data to upload payload data with a CSV file.
Next steps
Related topics
- For more information about debiased transactions, see Understanding how debiasing works.
Parent topic: Managing data for model evaluations in Watson OpenScale